Market Conditions for Housing Loans in Uganda

Market Conditions for Housing Loans in Uganda Annika Nilsson Introduction Housing is very capital intensive and constitute an important part of the development process. Housing finance markets are highly segmented in developing countries. In Uganda, less than four per cent of the population can afford a mortgage loan. However, microfinance clients may have access to loans that can be used to improve poor peoples housing. It is in the large gap between these two segments of the population that development of the mortgage market can help to provide better housing for the growing middle class. The majority of the population in Uganda, like most other developing countries, build their houses incrementally. They build over a period of several year with many small loans spread out over time. The interest rates on short term loans are higher than that of mortgages, making lending even more expensive than taking a mortgage loan. It is also time consuming and inefficient to build houses incrementally. Therefore, many houses never get finished. The high level of urbanization and lack of organized housing construction lead to increased slum populations. In Uganda the urban population is expected to double every 20 years (Statistical Abstract, 2010). In Kampala, the capital of Uganda, estimates of the growing housing deficit is one million units by 2025. Currently around 85 percent of the city population lives in slums. Despite this, in recent years the price of a well serviced house formal sector housing has skyrocketed (Giddings, 2009). The task for a developing country is to shift its housing finance system as far as possible towards the mortgage end of the spectrum and away from the incremental building end. A well functional financial system not only provide better and cheaper housing, but also contributes to growth (Levine, June 1997; Rajan & Zingales, June 1998). In The Mystery of Capital, the Peruvian economist Hernando de Soto suggests one explanation to why capital markets fail to reach out to the larger population; because the majority do not own formally titled property. The lack of formal titling thus prevents the use of property as collateral, and hence prevents the capital embedded in these assets from being unlocked (Soto, 2000). Other research support de Soto s in his view that less secure property rights are correlated to lower aggregate investment and slower economic growth (Knack & Keefer, November 1995; Mauro, August 1995; Svensson, July, 1998). 1

Is external finance, in addition to property rights, necessary for entrepreneurs to invest, or is property-rights security all that is needed? Previous research on cross-country studies cannot answer this question because effective protection for property rights is positively correlated with the use of external finance. For example, Rafael La Porta, Florencio Lopez-de-Silanes, Andrei Shleifer and Robert W. Vishny (1997, 1998, 2000) (La Porta, Silanes, Shleifer, & Vishny, July 1997) show more external finance is available when there is a stronger legal system in general and more effective protection of investors in particular, while Asli Demirguc-Kunt and Vojislav Maksimovic (1998) find that firms invest more from external funds in countries with secure property rights. According to the Uganda National Household Survey from 2005/2006 only half a percent of loans from formal institutions had a repayment period longer than three years, 45 percent had a repayment period of less than one year and 55 percent a repayment period of one to three years (Uganda National Household Survey, 2005/2006). The same survey shows that salary and land, 30 percent and 20 percent respectively, was the major forms of collateral in formal financial institutions (Uganda National Household Survey, 2005/2006). The question is thus why the majority prefer to take short term loans instead of long term loans that are cheaper. Previous research on barriers on the mortgage market mainly looked at the supply side of the mortgage market. Factors such as lack of valid land titles, difficulties in registering a mortgage and title transfer, and the difficulties creditors face in foreclosing on a property when a borrower defaults cite have been cited as reasons why the mortgage market fail to reach out to more borrowers (Butler, Krakova, & Safavian, 2009). Interest in the housing finance system as a whole is often limited to the question of how more resources can be made available for low-incoming housing. Studies on access to mortgage from the demand side has mostly been analyzed as a function of affordability, see for example How Low Can You Go: Charting the Housing Finance Access Frontier: A Review of Recent Demand and Supply Data (Meltzer, 2006). In this paper we go one step further by distinguishing the demand side on access and usage. Many more people may have access to a mortgage loan than those who choose to use it. While potential demand is usually observable, actual demand, in contrast, can be more difficult to understand. We look at the demand of mortgage loans among those applying for short term loans that will be used for investments in housing and land. Consequently, we focus on two different segments of the formal financial system, long term mortgage loans and short term loans which are either secured or not secured by the property. We will discuss the demand for mortgage as a function of its price and for a given set of state variables at the time of the survey. We are interested in the open question; could the group of short term borrowers be part of the long term mortgage market? Since our survey covers only persons taking short term loans, we cannot infer anything about the relative importance of mortgage loans for all persons building incrementally. We focus instead on persons that are already taking small loans for buying land and building a house. This paper have four parts. First we explain how the data was collected and the approach 2

of the survey. Then we investigate the demand and use of short term loans. We continue by investigatating the profile of borrowers. The reason is to figure out if this group can be eligible mortgage lenders (potential demand) and to get a better understanding about their use of financial resources for incremental building. Finally, we investigate the actual demand of mortgage loans. The Data Our data com from a survey of people borrowing money in one of the most popular banks in Uganda, Centenary Rural Bank. We chose Centenary Bank explicitly because it is the fifth largest bank in Uganda and it is applying a different way of screening clients and enforcing contracts compared to the four banks that are larger than Centenary Bank. The survey was carried out through a questionnaire during a period of three and a half weeks period in December-January 2010-2011. The persons selected for answering the questionnaire were applying for a home- or land loan at one of Centenary Banks branches in Kampala in December 2010. The branch of the bank in Kampala was randomly selected. During this time period all bank clients, i.e., 107 persons filled in the questionnaire. The final sample includes 67 persons, which corresponds to an answering frequency of 62 per cent. Reasons for not answering were divided into three categories; i) 11 persons completely refused without any reason, ii) 6 persons did not return the questionnaire, and iii) 23 persons did not have time to respond the big questionnaire. Centenary Rural Bank is offering two types of short term loans, home improvement loans and salary loans. Both can be used for housing and land. Therefore was those answering the questionnaire selected based on what the loan was intended to be used for, and not what kind of loan they were applying for. Centenary Bank was established in 1983 as the Centenary Rural Development Trust and was registered as a commercial bank in 1993, i.e., it is regulated by the Bank of Uganda. It is the fifth largest banks in Uganda. It offers both savings and loan products to all sectors of the Ugandan economy from microenterprises to small and medium enterprises (SMEs) and also to corporate clients. Centenary bank and Equity bank are the only commercial banks in Uganda with microfinance portfolios. There is limited information about their portfolios in micro-finance (Report on the Census of Micro-Finance Institutions in Uganda, 2011). Questions The design of the questionnaire incorporated both results of previous research, findings from previous interviews in Kampala and the result of a pilot study. The pilot study tested precisely how people understood various questions and established the best ways to ask in a easily understood way. For the purpose of this paper the most important issue was how to ask about land issues in a way that is understandably by the locals. The questions about land rights are also crosschecked 3

Real estates with high value & low volume Extension of formal property market Real estates with high volume & low value Financial barriers $ $ $ $ $ $ Existing Economy $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ Emerging Economy $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ Formal financial system Legal barriers Extension of formal financial market Informal sectors Figure 1: The real estate pyramide through several survey questions. We expected respondents would be reluctant to answer questions about their salary, and we therefore asked what level of education they have and what employment status they have. Previous survey shows that having a university degree and a regular job is positively correlated to a higher salary in urban Uganda (Uganda National Household Survey, 2005/2006) Approach The figure illustrates the amount of real estates in relations to their values. The top of the triangle illustrates properties that can be used for long term mortgage loans: houses with high value. The supply of housing in this high-end sector is comparatively small. In Uganda, less than one percent of housing construction is financed financed through the formal mortgage banking system. The bottom of the triangle contains the potential in currently market for short term loans in the formal and informal financial systems: far more people than in formal systems, but where the values of their properties are comparable low. Note, however, the sum value of the many small properties can exceed the total value of the properties currently contained within the formal systems. Indeed, one fact which emerges is that mortgage housing finance represents a very small portion of the portfolio of most financial institutions, and that the bottom of the pyramid, i.e., the lower income segments, are hardly served at all. 4

First tier: Mortgage loans Mortgage finance in Uganda is offered by: Housing Finance Bank, DFCU, Stanbic, Standard Chartered Bank and KCB. The range of mortgage products is varied. They are typically offered for up to 20 years in maturity and offered at variable rates. Interest rates are between 16 and 18 percent. Lenders ration credit because of risk and information problems. Credit risk is considered in terms of i) The borrowers ability to repay, ii) Collateral relative the value of the loan. Employment in the formal sector, public or private, should be verifiable and sustainable. i) The borrowers ability to repay Access to credit limited to those able to afford to make significant down payments; and able to pay less than a specific share of their income for repayments each month (35 per cent). There are other administrative costs involved when applying for a mortgage, such as: arrangement fees statutory fees, valuation fees, insurance, stamp duty, audited books of accounts incase the client operates a business and early repayment fees. ii) Collateral relative the value of the loan A standard loan to value (LTV) ceiling is 70 percent of the appraised value of the property. This because banks prefer to finance individuals who have completed what is locally known as a shell house (foundation, walls and roof completed). A maximum loan to value ratio is 80 percent for residential units in Kampala and 50 per cent for residential units in other towns. The property must have a valid title and building plans with local authority. The property must be situated in a well planned urban area. The plans must have water and electricity. And the construction must be in permanent materials. If the loan is for a house not situated in a urban area the borrower must have land situated in an urban area that can be pledged as collateral. In Uganda the value of the collateral is always higher than the loan size. Proceeds of the sale are distributed to clear the loan balance, auctioneers costs and any other costs partaining the property before the balance if any is credited on the mortagors account. Second tier: Centenary Rural Bank Centenary rural Development Bank, differ somewhat from other banks interviewed. Loans are given for a period up to three years. The bank give loan up to 80 per cent of the value of the loan. The interest rate for Home improvement loans is 22 percent. The bank do not ask for land as security for loans below 500 000 UGX. However, most loans are above 500, 000. The interest rate on salary loans is 19 percent also on reducing balance. The salary its self is the security. The borrower must also show that he or she had a well-managed account with the bank or with some other bank for some time. These lending requirements are easier to meet for small customers than the requirements of other commercial banks. 5

Borrowers that have problems in repaying their loan get extra time to fulfill their repayment of the loan. Thus the repayment ratio is nearly hundred percent and the bank never sell collateral in order to recover the loan. This bank screens loan applicants by requiring them to have a business that has been on-going for at least one year, a viable business plan and regular cash flow. collateral is also required, but can be either in movable or immovable assets. To be able to monitor the borrower s business, Centenary Rural Development Bank also requires the business to be situated within 30 km from the closest bank branch. Short term loans for incremental building In order to get a better understanding about incremental building we raised several questions related to Frequency of taking short-term loans at Centenary Bank: For 50 percent of those participating in the survey is it the first time they apply for this kind of loan at Centenary Bank. Reasons for taking loans at Centenary Bank: In order to get a better understanding about what the loan will be used for we asked what the applicant was going to use the loan for. More than half of respondents among first time lenders and those that had taken a loan before said that the loan was taken to buy land. Women is to a larger extent than men going to use the loan for land. This may be explained by the fact that for 61 per cent of women is this their first application, compared to 40 per cent of men. Among those that had taken a loan before where the level of buying land among men higher than among women, 76 per cent and 50 per cent respectively. The second largest reason for applying for group a loan in order to finish their house. This might be explained by the fact that they can move into the house when the house is finished. To summarize, the data shows that most loans are taken for covering the cost of land, both for those taking their first loan as well as those that took a loan before, even when comparing the whole sample with those having a regular job and especially for women. Since women only hold 9 percent of registered land in Uganda (Ellis, Manuel, & Blackden, 2006) do these results shows that ownership of land among is increasing substantially with the help of loans. Other financial sources: Applicants were asked what other financial sources they had used apart from short term loans from Centenary bank. The majority, around 60 per cent, answered that apart from taking or applying for short term loans they used savings to finance their land and house. How much of the cost would you like to be covered by short term loans: We asked how much of the cost of the house they would like to be covered by loans. The majority, 42 per cent, would like the loan to cover 30-50 per cent of the whole cost of the house, the second majority, 37 per cent, would like to cover the cost with loans equivalent to 10-30 per cent of the final cost and the third majority, 9 percent, want to finance their house with as much 6

as 70-100 per cent. Those having a regular job and title, i.e., those that most probably would qualify for a mortgage loan do to a slightly higher degree desiring loans to cover the final cost of their house. How many years will it take to finish the house: The respondents were further asked how many years they think it will take to finish the house. The majority, 47 per cent, answered that it will take 1-3 years. Followed by the group, 32 per cent, that estimate it will take less than one year, 32 per cent. And thirdly the group that don t know represents 18 per cent. To conclude, we see that loans are mainly used for buying land and secondly to finish a house. Buying land requires a large sum, which may explain why most loans are taken to buy land. Apart from short term loans savings are mainly used. Women is to a larger extent than men going to use the loan for land. Potential demand for mortgage loans The borrowers profile is surveyed through several questions. This in order to see if this group could be eligible for mortgage loans, i.e., we are investigating the potential demand for mortgages. We also investigated the level of women participating in the survey and possible gender differences. Gender: There is a almost equal participation of men and women in this survey: 40 per cent are women. The response rate for this question is 100 percent. This in contrast to a a World Bank study suggesting that women only receive 9 percent of all credit in Uganda (Ellis et al., 2006). However, one has to consider that women in urban areas have a higher level of education and formal employment than women in rural areas. Age: The majority in the survey, 37 per cent, is between 26 to 35 years old, followed by the group in the age of 36-45 that (24 per cent) and the third group, 18 per cent, is between 46 to 55 old. Education and employment: Since we expected respondents reluctant to answer correctly about their formal income we instead asked about their education level and their status of employment, since these two parameters are intercorrelated. The majority, 83 per cent, have a University degree. Of those having a University degree the major part, 66 per cent, have a bachelor degree and 16 per cent have a masters degree. Twelve percent, have a s6 education level. Men and women have a similar level of University education as the average of the sample, 81 and 84 per cent respectively. Not surprisingly is the level of education and employment correlated. The majority, 72 per cent have a regular job that is paid weekly or monthly, 18 per cent, is self employed and the third largest group, 15 per cent, is self employed with their own company. The difference between that you are self employed and that you are self employed with your own company is that the 7

latter definition means that you have a legal registered company with audited accounts, i.e., belongs to the formal economy. Of those regularly employed is almost half of them women, 47 per cent, indicating that women are regularly employed to a higher degree than men, 85 and 64 per cent respectively. The response rate for these questions is 100 percent. Land ownership: Collateral in form of registered land title is a requisite to get a long term mortgage loan. In order to see if the sample in this survey could be eligible for a mortgage loan we raised several questions related to the requirements of mortgage loans. Asked whether they owns land since before more than the majority answered yes. Affirmatively ranged from 62 per cent of the whole group, 68 per of men, 54 per cent of women and those having a regular job. These answers are naturally correlated to that the average of the whole group and men had taken more loans before compared to women and those regularly employed. Since only land in urban areas is acceptable as collateral for a long term mortgage loan we further asked where the land they own is situated. All answered that the land they owns is situated in surrounding of Kampala. Only registered land in Land Ministry is accepted as collateral for a mortgage loan. Half, 49 percent, of those owning land since before answered that they have a land title registered in Land Ministry, and 24 per cent answered that their title is on process to get registered. However, by crosschecking the answers does it seem that those answering that their title is on process have not started the process yet. One possible interpretation of those answering that that their title is on process is that they are considering registering it within a near future. A high portion, 22 percent, answered that they have Kibanja land. Kibanja land is accepted as collateral at Centenary Bank, although the interest charged is around ten per cent higher than for a formal registered land title. Kibanja land is not accepted as collateral for a long term mortgage loan. Kibanja land is the land occupied by a squatter. If a piece of land has a title but is occupied by a squatter, it is the latter s s and the titleholder s registered land. Both the squatter (at holder) and the titleholder (landlord) have interests in the same piece of land and neither can remove the other at will. The problem arises when a squatter wants to sell his age loan. The Land Act requires that the Kibanja holder not only informs the landlord of the intention to sell but also gives him the first priority to buy. The response rates for these questions range from 95 to 100 percent. In Uganda there are four different tenure systems. Therefore, we asked those having a title under what land tenure their land falls. The survey shows that the majority have land that is acceptable for a long term mortgage loan. As much as 100 per cent of the total sample and of those having a regular job have mailo land. There are two different types of mailo land. Both of them are accpeted as colalteral for a mortgage loan. The response rate for this question is 100 percent both for the total sample and those having a regular job. To summarize, the majority of this group could theoretically be eligible for a mortgage loan since they have a regular job and land that is accepted as collateral for long term mortgage loan. The survey also shows that almost half of respondents are women and that women to a higher 8

Table 1: Title on land Frequency (%) Total Regular job Women Men Has a certificate on title 48 44 36 52 Title is on process 25 24 29 24 It is Kibanja land 22 28 36 16 I do not know 5 4 0 8 Number of respondents 40 25 14 25 Skipped to answer 0 2 0 0 Table 2: What kind of land tenure Frequency (%) Total Regular job Private mailo 85 83 Bulange (official mailo) 15 17 Freehold 0 0 Leasehold 0 0 I do not know 0 0 Number of respondents 40 12 Skipped to answer 0 0 9

degree than men have a regular job. Women are to a lower extent than men owning land since before, which could be related to that men had taken more loans before. Actual demand for long term mortgages In order to get a better understanding about the actual demand for mortgage loans we asked the respondents if they either have applied or within three years will apply for a larger long term mortgage loan. Only 5 per cent answered that they will apply for a mortgage loan, the majority, 54 per cent, answered that they will not apply for a mortgage loan, and the third largest group, 41 per cent, answered that they do not know. The pattern of answers are similar for those having a regular job as well as for women and for men. The response rate was 94 per cent. Table 3: Have you applied before or will you within three years apply for a mortgage loan Frequency (%) Total sample Regular job Women Men YES 5 7 8 3 NO 54 58 54 54 I do not know 41 36 37 43 Answered the question 63 45 24 37 Skipped to answer 4 3 2 2 The group that answered that they not have applied for a mortgage loan neither want to do it within three years were further asked to give reasons why. See Table 4. For the majority, 54 per cent, is the main reason why they haven t applied for a mortgage, neither will apply, is because it is too complicated. Too complicated: In order to take a mortgage loan the bank requires securities such as identification, spouse consent, valid title, insurance, proof of formal income, and valuation of the property. Mortgage lending is also, like in many developing countries constrained by lack of formal registration systems for property(giddings, 2009). Registering property in Uganda requires 13 procedures, takes 77 days and costs 3.2 percent of the property value. Globally, in 2010 Uganda stand at 155 in the ranking of 183 economies in terms of the cost, number of steps and time to register a property (The International Bank for Reconstruction and Development, 2012). Too low income: The second majority, 28 per cent, answered that their income was too low for applying for a mortgage loan. This reason for those answering that their income is too low is rather lack of access than self exclusion. Of those indicating that their salary is too low 92 percent have a regular job and 58 have a University degree. Since only four percent of the population can afford a mortgage loan, these results show that the income requirement for mortgage loans leave the 10

middle class outside the mortgage market. Afraid of foreclosure process: The third largest reason, 19 per cent of respondents, do not want to apply for a mortgage because they fear they will loose their land title if they cannot pay back. In Uganda, if one fail to pay back the loan the collateral becomes sold after six months. Since the value of the collateral is often higher than the loan amount and also constitute their actual home, many are afraid of loosing their loan. In contrast to Centenary Bank which gives the client extra time to fulfill repayments in case of difficulties. This shows that inadequate enforcement methods make borrowers reluctant to take long term loans. Other reason: A high portion answered other reason that they explained with factors such as; they do not want a long term loan due to insecure labor market, do not want to be indebted for a long time, and that they are generally scared of banks. Lack of collateral: Lack of land to use as security is only indicated by 5 per cent of respondents as the reason why they do not apply for a mortgage loan. Table 4: Why haven t you applied neither want to apply for a mortgage loan within three years Frequency (%) Total sample Regular job Women Men I lack acceptable land as security 3 4 8 0 My income is too low 27 36 33 25 I am afraid of loosing title if I can t pay back 24 24 25 25 Afraid of loosing title, corruption Land Ministry 0 0 0 0 Too complicated 61 60 42 70 Bad credit history 0 0 0 0 Other reason 21 16 42 10 Answered the question 33 26 12 20 Skipped to answer 1 0 1 0 Conclusions Only four per cent of Ugandans can afford a mortgage loan. The majority of houses are therefore built incrementally with short term loans. In this paper we present a survey about bank clients applying for short term loans in Centenary Bank in Uganda. The purpose was to investigate whether this group of lenders could be eligible for a larger mortgage loan. The survey was given to all customers who were seeking any type of financing through the bank for either for building a house or buying land in Kampala. The survey shows that the majority of those applying for short term loan use the loan to buy land, and secondly to finish the house over a period of some years. Apart from using short-term loans do they use savings to finance their house and land. 11

Furthermore, the survey shows that the majority of respondents represent a rather homogenous group of a young middle class. Furthermore, the survey shows that access to credit increases women s ownership of land. Three quarters of this group are in their thirties and formally employed. Forty per cent is women. Women have a bachelor degree and formal employment to a higher degree than men. They are also going to use the loan to buy land to a higher degree than men. One reason is that over 60 per cent of women apply for a loan for the first time, which they usually need for buying land, compared to 40 per cent of men. These figures stand in large contrast to that currently only seven percent of women in Uganda own land. The main reasons why this group do not want to take cheaper mortgage loans are that they are too complicated to apply for, their income is to low to qualify for such loans, and they are afraid of loosing their collateral if they would be unable to meet their repayments. Rather than taking a less expensive loan do they therefore prefer to combine several shorter term loans due to the lack of functional regulation for mortgage lending. These results tell us that there is enormous scope for improving intermediation of mortgages - at least as far as the middle-income are concerned. They have a stable income and are able and willing to pay much higher rates than for mortgages. If there was some way to improve credit delivery models to allow these banks to lend more to them, both sides could greatly benefit. 12

References Butler, S., Krakova, M., & Safavian, M. (2009). Mortgage Registration and foreclosure around the Globe: Evidence from 42 Countries. Housing Finance International, September, 19-43. Ellis, A., Manuel, C., & Blackden, C. M. (2006). Gender and Economic Growth in Uganda - Unleashing the power of women (W. B. Publication, Ed.). The World Bank. Giddings, W. S. (2009). The land market in kampala, uganda and its effect on settlement patterns. International Housing Coalition. Knack, S., & Keefer, P. (November 1995). Institutions and economic performance: Cross country tests using alternative institutional measures. Economics and Politics, 7 (3), 207-228. La Porta, R., Silanes, F. Lopez-de, Shleifer, A., & Vishny, R. (July 1997). Legal determinants of external finance. Journal of Finance, 52 (3), 1131-1150. Levine, R. (June 1997). Financial development and economic growth: Views and agenda. Journal of Economic Literature, 35 (2), 688-726. Mauro, P. (August 1995). Corruption and growth. Quartely Journal of Economics, 110 (3), 681-712. Meltzer, I. (2006). How Low Can You Go: Charting the Housing Finance Access Frontier: A Review of Recent Demand and Supply Data (Tech. Rep.). Centre for Affordable Housing Finance in Africa, a division of FinMark Trust. Rajan, R., & Zingales, L. (June 1998). Financial dependence and growth. American Economic Review, 88(3), 559-586. Report on the Census of Micro-Finance Institutions in Uganda. (2011, June). Soto, H. de. (2000). The mystery of capital why capitalism triumphs in the west and fails everywhere else. Basic Books. Statistical Abstract. (2010). The Republic of Uganda: The Ministry of Lands, Housing and Urban Development. Svensson, J. (July, 1998). Investment, property rights and political instability. European Economic Review, 42(7), 1317-41. The International Bank for Reconstruction and Development. (2012). Doing Business in a more trasparent world - Economy Profile Uganda. The World Bank. Uganda National Household Survey. (2005/2006). Uganda Bureau of Statistics. 13